Assessment of Sedentary Behavior With the International
Physical Activity Questionnaire
Rosenberg, D. E., Bull, F., Marshall, A. L., Sallis, J. F., & Bauman, A. E. (2008). Assessment of Sedentary
Behavior With the International Physical Activity Questionnaire. Journal of Physical Activity & Health,
5(Supp. 1), S30-S44.
Published in:
Journal of Physical Activity & Health
Document Version
Publisher's PDF, also known as Version of record
Link to publication in the UWA Research Repository
Rights statement
Link to Publisher's website supplied in Alternative Location.
General rights
Copyright owners retain the copyright for their material stored in the UWA Research Repository. The University grants no end-user
rights beyond those which are provided by the Australian Copyright Act 1968. Users may make use of the material in the Repository
providing due attribution is given and the use is in accordance with the Copyright Act 1968.
Take down policy
If you believe this document infringes copyright, raise a complaint by contacting [email protected]. The document will be
immediately withdrawn from public access while the complaint is being investigated.
Download date: 15. Jun. 2017
Journal of Physical Activity & Health, 2008, 5, Supp 1, S30-S44
© 2008 Human Kinetics, Inc.
Assessment of Sedentary Behavior
With the International Physical Activity
Questionnaire
Dori E. Rosenberg, Fiona C. Bull, Alison L. Marshall,
James F. Sallis, and Adrian E. Bauman
Purpose: This study explored definitions of sedentary behavior and examined
the relationship between sitting time and physical inactivity using the sitting
items from the International Physical Activity Questionnaire (IPAQ). Methods:
Participants (N = 289, 44.6% male, mean age = 35.93) from 3 countries completed
self-administered long- and short-IPAQ sitting items. Participants wore accelero­
meters; were classified as inactive (no leisure-time activity), insufficiently active,
or meeting recommendations; and were classified into tertiles of sitting behavior.
Results: Reliability of sitting time was acceptable for men and women. Correlations between total sitting and accelerometer counts/min <100 were significant
for both long (r = .33) and short (r = .34) forms. There was no agreement between
tertiles of sitting and the inactivity category (kappa = .02, P = .68). Conclusion:
Sedentary behavior should be explicitly measured in population surveillance and
research instead of being defined by lack of physical activity.
Keywords: health promotion, physical inactivity, health behavior, exercise
Physical inactivity and sedentary behavior are cited as primary contributors to
the obesity epidemic.1 Though leisure-time physical activity levels have changed
little over the past few decades,2,3 sedentary behaviors such as television watching
and use of motorized transport have increased and, therefore, might help explain
the increase in obesity.4 Research on sedentary behavior is a public health priority,
and thoughtful conceptualization and measurement are precursors to high-quality
research.
A major shortcoming of previous research on sedentary behavior is that
researchers have not defined and measured sedentary behavior consistently. A
common problem is a lack of clarity in referring to sedentary behavior and sedentary
Rosenberg is with the Joint Doctoral Program in Clinical Psychology, San Diego State University and
University of California, San Diego, CA 92103. Bull is with the School of Sport and Exercise Sciences,
Loughborough University, Leicesteshire, LE11 3TU, U.K. Marshall is with the School of Public Health,
Queensland Institute of Technology, Kelvin Grove, QLD 4059, Australia. Sallis is with the Psychology
Dept, San Diego State University, San Diego, CA 92182. Bauman is with the School of Public Health,
University of Sydney, NSW 2006, Australia.
S30
IPAQ Sedentary Behavior S31
people, who are often identified as having low levels of physical activity. TudorLocke and Meyers5 explain:
To date, sedentarism has been inferred according to comparatively low levels
of total energy expenditure, time or distance walked, stairs climbed, and/or
through lack of self-reported participation in vigorous leisure activities, including sports and exercise.(p 92)
Healthy People 2010 states that sedentary person “denotes a person who is relatively inactive and has a lifestyle characterized by a lot of sitting.”6(p 22-36) Some
researchers have suggested that individuals spending less than 10% of their daily
energy expenditure in moderate- to high-intensity activities should be classified
as sedentary.7,8
Some of the confusion that arises from the different definitions might be caused
by defining sedentary individuals as those who are inactive or not physically active
as opposed to those who spend a great deal of time doing sedentary behaviors. It
is inadequate, however, to infer sedentary behavior from lack of physical activity
because there is ample evidence that they are independent behaviors that have different effects on health.3,5,9-12 Thus, defining sedentary people as having deficient
levels of physical activity is inaccurate and can confuse attempts to measure the
sedentary-behavior construct. We support referring to people with low levels of
physical activity as insufficiently active rather than sedentary.13 Biddle et al assert
that sedentary behavior is “a distinct class of behaviors characterized by low energy
expenditure.”12(p 30) This latter definition is more specific and suggests a separate set
of behavior domains than presence or absence of physical activities.
Sedentary behavior in adults is an area that can draw on extensive literature on
youth. The predominant sedentary behavior studied in children is television viewing. It has been hypothesized that sedentary behaviors such as television watching
displace engagement in physical activities and lead to obesity.12 There is evidence
that television watching stimulates the consumption of unhealthful foods.14,15
A meta-analysis16 found significant negative relationships between television
and computer/video-game use and body fatness in children and youth, although the
effect sizes were small. Relationships of television viewing and computer/videogame use to physical activity were small and negative. Cluster analyses of sedentary
behavior support the conclusion that television viewing does not substantially displace physical activity among youth. Youth can report high levels of both physical
activity and sedentary pursuits.17,18
Studies of adults also have focused on television viewing as a means of examining sedentary behavior. The average US adult spent 29 hours per week watching
television in 2003.19 Cross-sectional and prospective studies generally showed
obesity rates 1.5 to 2 times higher among those who watched the most television,
even after adjustment for physical activity.19-31 Some studies have shown mixed
results in which relationships were only significant for certain age groups27,29 or
only cross-sectionally.30
More recently, researchers have measured a wider range of sedentary behaviors.
There is less literature that is based on questions about time spent sitting.19,20,32,33
The limitations in this set of literature are that the measures have been brief, some
focused only on recreational sitting time, and most did not report any reliability
or validity data. One study reported moderately high reliability correlations for
S32 Rosenberg et al
items from the Western Australian Incidental Physical Activity Questionnaire that
assessed time spent sitting and higher coefficients for items measuring computer
and television use.34 Unfortunately, there is no consistency in the way sitting time
is measured, so studies cannot be compared. However, all of these studies reported
a positive relationship between sitting time and weight status.19,20,32,33
Population-surveillance surveys aim to measure indicators of sedentarism in
large populations and do not comprehensively assess health behaviors. Population surveillance in adults has predominantly defined sedentary lifestyles based
on lack of physical activity. For example, the Behavioral Risk Factor Surveillance
System (BRFSS) classifies people as inactive if they do not engage in leisure-time
activities.35-37 Such classification methods do not directly address the sedentary
behaviors that would contribute to inactivity,9 but at present, they are the primary
source of population estimates of sedentary behavior in the United States. Healthy
People 2010 states that “the message that a sedentary lifestyle plays a role in both
overweight and weight loss needs to be addressed better.”6(p 22-6)
There are 2 important problems with the measurement of sedentary behavior.
One is the lack of consistency with terminology and confusing definitions of sedentary individuals and sedentary behaviors. Second, there is an overreliance on
using television viewing as a proxy measure of sedentary behavior. Thus, there is
a need for a brief, yet reliable, and valid measure of sedentary behavior that can be
included in population-monitoring surveys so that the independent contributions of
sedentary behaviors to obesity and other health outcomes can be determined.
The purpose of the present study was to extend previous reports38 of the reliability and validity of the sitting item from the International Physical Activity
Questionnaire (IPAQ) to assess measurement properties by gender. To evaluate the
adequacy of using lack of physical activity during leisure time as the definition of
sedentarism, the second aim was to compare the relationship of sedentary behavior
measured by IPAQ’s sitting item with inactivity measured using IPAQ’s physical
activity questions scored to reflect Healthy People 2010 definitions of inactivity.
Methods
Procedure
The reliability and validity of the IPAQ was originally tested in 12 countries with
a total sample exceeding 2000.37 Eight IPAQ versions were tested. In the original
reliability and validity study, the authors concluded that the last-7-days time frame
was recommended.38 The site subsamples included in the present study were
selected because they used the same version of the IPAQ and had validity data.
Thus, included sites had to have administered both the short and long IPAQ versions that used the recommended past-week reference period, and the IPAQ had
to have been self-administered. This resulted in the inclusion of data from 4 sites
across 3 countries.
Data Collection
Data from the original IPAQ reliability and validity study sites—Bristol (UK), San
Diego (USA 1), South Carolina (USA 2), and the Netherlands—were included in
IPAQ Sedentary Behavior S33
this study. All sites used the English version of the IPAQ except the Netherlands
sample, which administered a Dutch version. Although most IPAQ versions have
been translated and back-translated into English, the Dutch version was not backtranslated, but it followed the cultural-adaptation guidelines described in Craig et
al.37 All versions are available at http://www.ipaq.ki.se/ipaq.htm. Most sites used
convenience samples consisting of staff, students, or others affiliated with universities. These sites’ samples had generally high education and socioeconomic status
(see Table 1). More detailed information has been previously presented.38
Data were collected at all sites following a standard protocol for administering
the questionnaire on 3 occasions. During visit 1, participant demographic information was collected, consent was given, and the accelerometer was initialized and
given to the participant to wear for 1 week. At visit 2, 1 week after visit 1, participants returned the accelerometers and each completed an IPAQ. The IPAQ was
readministered 3 to 7 days later (visit 3) to allow overlapping days for test–retest
analyses.
Measures
Sitting Items. The sitting items in the long and short versions of the IPAQ last-7-
days version were similar. For both versions, participants were instructed to think
about the time they spent sitting at work, at home, while doing course work, and
during leisure time. They were asked to estimate in total the number of hours and
minutes per day they spent sitting for a weekday and a weekend day. For IPAQ long
forms, however, participants were instructed not to include time spent sitting in a
motor vehicle. Time spent in motorized transportation was assessed separately on
the long form only. The exact item wording is available at www.ipaq.ki.se.
Objective Measure of Sedentary Time. The Computer Science and Applica-
tion Inc’s (Shalimar, FL) accelerometers (CSA model 7164, now available from
www.theactigraph.com) were used to objectively assess participants’ sedentary
time. Participants were instructed to wear the accelerometer during waking hours
for 7 days. Accelerometer data were included in analyses if there were at least
600 minutes of time recorded each day for at least 5 days, 1 of which had to be a
weekend day.37 Data were stored in 1-minute intervals. Minutes with counts <100
per minute were considered to represent time spent in sedentary pursuits.39 The use
of counts <100 per minute captures time spent sitting or lying still such as while
watching television and playing video games39-42 and has been used as the cut point
for sedentary behavior in previous studies with youth.39,43 Using this cutoff helps
distinguish between engaging in sedentary activities and light activities (eg, ironing, washing dishes, sweeping the floor, and cooking), which have been measured
at 150 to 400 counts per minute.44,45
Measure of Physical Inactivity and Recommended Physical Activity. Definitions
of inactivity and recommended physical activity used in Healthy People 2010 6 and
US national surveillance data36 were approximated with data from IPAQ items. The
BRFSS used in US national surveillance asks individuals to take account of recreational activity and gardening and yard activities, and people who report no such
activities are classified as inactive. In the present study, based on appropriate IPAQ
items, individuals were classified as meeting physical activity recommendations
S34
157 (48.4)
74 (45.9)
28 (25)
30 (46.7)
289 (45.1)
Site
UK
Netherlands
USA 1
USA 2
Total sample
35.9 (11.3)
36.1 (12.6)
48.9 (6.1)
32.7 (10.9)
35.3 (10.4)
Mean age
in y (SD)
16.3 (3.4)
17.3 (3.0)
—c
—b
16.3 (3.6)
Mean y of
education
(SD)
48.69 (20.4), 56.03 (23.52)
37.86 (20.78), 53.01 (19.8)
47.54 (26.11), 64.33 (27.3)
55.49(19.4), 60.04(21.0)
46.45 (19.2), 53.80 (23.0)
Mean h/wk of total sitting
and total sitting including
transportation (SD)a
b
a
Derived from the International Physical Activity Questionnaire (IPAQ) long form.
70% of the Netherlands sample had >18 y of education.38
c
Education level not available for USA 1 sample.
Gender, N
(% male)
Table 1 Sample Characteristics
53.02 (9.5), N = 212
53.13 (9.8), N = 26
48.88 (10.1), N = 26
55.55 (10.5), N = 34
53.17 (8.8), N = 126
Mean h/wk
accelerometer counts
<100 (SD)
8.6, 41.7, 49.6
6.7, 30, 63.3
7.4, 29.6, 63
3, 48.5, 48.5
11.1, 45.1, 43.8
% inactive, insufficiently
active, active during
leisure time
IPAQ Sedentary Behavior S35
(30 minutes of moderate recreational, walking, or gardening activity 5 days per week
or 20 minutes of vigorous recreational or gardening activity 3 days per week), being
insufficiently active (not meeting recommendations but not completely inactive), or
inactive (no moderate or vigorous recreational or gardening activities).
Analysis
Analysis occurred in 3 parts. First, test–retest reliability of the sitting items was
examined separately for men and women for each of the following definitions of
sitting. For the long form, test–retest correlations were conducted for weekday,
weekend, weekday + weekend (total), transportation sitting, and transport + weekday + weekend (grand total) sitting. For the short form, test–retest correlations were
conducted for weekday, weekend, and weekday + weekend (total) sitting. Retests
(at visit 3) were performed 3 to 7 days after the initial test at visit 2. Participants
who completed the IPAQ administration more than 8 days after the initial test were
excluded from the analyses. Spearman correlations were used because data were
nonnormally distributed.
Next, to assess criterion validity of the sitting item by gender, total time spent
sitting was correlated with the number of accelerometer counts <100 per minute.
For the long form, total sitting and grand-total sitting (which includes sitting for
transportation) were correlated with accelerometer counts <100 per minute. For
the short form, total sitting was correlated with accelerometer counts <100 per
minute.
Finally, to assess the agreement between classification of individuals identified
as inactive during leisure time and those identified as spending high amounts of
time sitting, kappa statistics were run. Participants’ classification as inactive, insufficiently active, or meeting physical activity guidelines was entered as 1 variable.
Participants were categorized into tertiles of sitting behavior based on their responses
to the IPAQ long-version sitting items, and this was entered as the second variable.
Kappa analyses were conducted only for the long form because it contained more
specific domains that allowed gardening and yard activity to be taken into account
separately from recreational activities to replicate BRFSS criteria.
Results
Demographics
Participant characteristics are presented in Table 1. There were 28 to 157 participants
per site. All sites had more women than men participate in the study. In general,
the participants reported high levels of education.
Test–Retest Reliability
Test–retest Spearman correlations for the long form were very good for men and
women (Table 2). Reliability coefficients were similar for the different components
of sitting—weekday, weekend, total, transport, and grand-total sitting. Men and
women had similar reliability coefficients on all sitting items. The coefficients
ranged from .40 (USA 2, time spent sitting on the weekend) to 1.0 (USA 1, time
S36 Rosenberg et al
Table 2 Test–Retest Reliability Coefficients for the Sitting Items
of the IPAQ Long Form
Sample
UK
men
women
all
Netherlands
men
women
all
USA 1
men
women
all
USA 2
men
women
all
Total
men
women
all
N
Weekday Weekend
Totala Transport (N)b
Grand totalc
(N)b
65
78
145
.81
.64
.72
.78
.79
.79
.83
.66
.75
.76 (68)
.85 (79)
.81 (149)
.82 (68)
.65 (79)
.74 (149)
28
38
66
.96
.96
.96
.95
.88
.91
.97
.93
.96
.95 (29)
.94 (39)
.93 (68)
.78 (29)
.93 (39)
.87 (68)
7
18
25
1.0
.92
.95
1.0
.95
.97
1.0
.94
.96
13
16
29
.91
.82
.82
.40
.90
.78
.91
.86
.87
.85 (14)
.93
.91
.89 (14)
.88
.85
109
144
255
.86
.77
.81
.82
.85
.84
.87
.78
.82
.80 (113)
.87 (146)
.84 (261)
.83 (113)
.77 (146)
.81 (261)
.96
.83
.84
1.0
.91
.95
Abbreviation: IPAQ, International Physical Activity Questionnaire.
a
Total sitting = weekday + weekend sitting.
b
Sample sizes that differ from the sample sizes in column 2 are reported in parentheses.
c
Grand-total sitting = total + transportation sitting.
spent sitting on a weekday, weekend, and total sitting) for men and from .64
(UK, time spent sitting on a weekday) to .96 (Netherlands, time spent sitting on
a weekday) for women. Most of the correlations were above .75, indicating very
good to excellent test–retest reliability for the sitting items in the IPAQ past-7-days
long form.
Test–retest reliabilities for the short form were also in acceptable ranges for
men and women (Table 3). Coefficients were slightly lower than those observed for
the long form for weekday and weekend sitting but similar for total sitting. Again,
men and women tended to have similar test–retest correlations. Correlations ranged
from .77 (UK, weekday sitting) to .98 (USA 1, weekend sitting) for men and from
.62 (UK, weekday sitting) to .96 (Netherlands, weekday sitting) for women. Most
of the correlations were above .70, again indicating acceptable test–retest reliability
for the sitting items in the IPAQ past-7-days short form.
IPAQ Sedentary Behavior S37
Table 3 Test–Retest Reliability Coefficients for the Sitting Items
of the IPAQ Short Form
Sample
UK
men
women
all
Netherlands
men
women
all
USA 1
men
women
all
USA 2
men
women
all
Total
men
women
all
N
Weekday
Weekend (N)a
Totalb (N)a
65
79
146
.79
.62
.70
.77 (64)
.71 (80)
.75
.81
.63 (80)
.73 (147)
25
39
64
.94
.96
.96
.90
.92
.91
.93
.94
.95
7
21
28
.86
.96
.93
.98
.85
.87
.79
.94
.92
14
16
30
.83
.77
.79
.79
.85
.84
.86
.90
.85
106
149
257
.61
.58
.59
.70 (105)
.73 (150)
.72
.84
.77 (150)
.81 (258)
Abbreviation: IPAQ, International Physical Activity Questionnaire.
a
Sample sizes that differ from the sample sizes in column 2 are reported in parentheses.
b
Total sitting = weekday + weekend sitting.
Criterion Validity
Spearman correlation coefficients between accelerometer counts and sitting time
reflected low to moderate agreement in general (Table 4). All correlations were positive except among men in the Netherlands sample, in which a negative correlation
with accelerometer counts was observed. Overall, women had higher correlations
than men. Data about sitting from the short and long forms performed similarly
against the accelerometer estimates of sedentary behavior. Most correlations were
above .25.
Agreement With Guidelines
Figure 1 shows the relationship between tertiles of self-reported sitting and the cut
points for physical inactivity, insufficient activity, and meeting recommendations.
The kappa statistic was not significant, indicating poor agreement between the
tertiles of time spent sitting and various categories of activity.
S38 Rosenberg et al
Table 4 Spearman Correlation Coefficients for Sitting Items With
Accelerometer Counts <100 per min for Long and Short IPAQ Forms
Sample
UK
men
women
all
Netherlands
men
women
all
USA 1
men
women
all
USA 2
men
women
all
Total
men
women
all
N
Long form
Long form +
transportation (N)a
Short form (N)a
56
61
118
.24
.28
.24
.22 (58)
.35
.25 (120)
.24
.29 (62)
.25 (119)
11
19
30
–.16
.44
.26
–.16
.56
.35
6
20
26
.31
.41
.30
.37
.30
.26
.37
.52
.45
13
13
26
.60
.37
.50
.63
.39
.49
.59
.48
.49
86
113
200
.26
.40
.33
.23 (88)
.38
.31 (202)
.24 (84)
.43 (114)
.34 (199)
–.48 (9)
.43
.22 (28)
Abbreviation: IPAQ, International Physical Activity Questionnaire.
a
Sample sizes that differ from the sample sizes in column 2 are reported in parentheses.
Discussion
The current study assessed measurement properties of the IPAQ sitting items and
explored current definitions of sedentary behavior and inactivity. In general, the
sitting items from the last-7-days IPAQ were reliable and valid among participants
from several countries. The differences between the reliability coefficients for men
and women were small, thus demonstrating that the sitting items were reliable to
use in both populations.
In this study, the correlations between the IPAQ sitting time and accelerometer
counts depicting sedentary behavior (<100 counts/min) were small to medium. The
correlations were comparable to the small but significant relationships reported
between multiple self-reported measures of physical activity and accelerometers.46
Thus, reported sitting is approximately as valid as reported physical activity. There
was, however, a notable unexpected finding of negative associations between
reported sitting and accelerometer sedentary time for men from the Netherlands.
There could be a few explanations for this anomalous finding. The IPAQ in the
Netherlands was administered in Dutch, but all other versions were administered
in English. Because the Dutch version was not back-translated, there is the possibility of errors in translation. In addition, individuals in the Netherlands tend to
IPAQ Sedentary Behavior S39
Figure 1 — Agreement between tertiles of sitting and physical inactivity categories based
on the IPAQ long form for the total sample. Note. Sample divided into high, middle, or low
sitting time derived from the IPAQ long-form sitting item. Sample categorized as active
(meeting physical activity recommendation—30 minutes of moderate recreational, walking,
or gardening activity 5 days per week or 20 minutes of vigorous recreational or gardening
activity 3 days per week), insufficiently active (not meeting activity recommendation but
not completely inactive), or inactive (no moderate or vigorous recreational or gardening
activities).
engage in high amounts of biking, which is not measured by accelerometers and
could be counted as sedentary time.
The test–retest reliability for all of the sitting items was >.75, indicating high
reliability for a self-reported measure. For a brief measure, the IPAQ sitting items
have adequate reliability and validity to use for surveillance. When the focus of
research is to improve understanding of sedentary behavior specifically, a more
detailed measure should be used. Except for men from the Netherlands, reliability
and validity of the IPAQ sitting items appear to be generalizable to men and women
in multiple countries.
Describing individuals who are inactive during leisure time as sedentary was
shown to be a misnomer. Although it is true that the concepts of inactivity and
sedentary behavior are conceptually related, these behaviors are empirically distinct
and should be considered as such both in measurement methods and the terms
used to define them. Using these terms interchangeably causes ambiguity. Data
from the present study demonstrated the lack of agreement between what can be
characterized as sedentary behavior (sitting time) and what can be characterized as
physical inactivity. Participants in this study who reported high amounts of sitting
were just as likely to be classified as active on the physical activity measure as
those who reported sitting the least. Among the highest tertile of sitters, 44% were
classified as meeting physical activity guidelines. This suggests that individuals
who are sedentary for large amounts of time can also engage in the recommended
S40 Rosenberg et al
levels of leisure-time activity; thus, the 2 behaviors are not mutually exclusive. It
is possible that individuals who sit for lengthy amounts of time at work and watch
a lot of television might still go to the gym for 30 minutes a few times per week
and, thus, be considered active. These individuals might meet physical activity
recommendations, but their high levels of sedentary behavior throughout the day
could be independently associated with health problems.
Accurate and reliable measures are needed specifically to assess sedentary
behavior so that associations with health outcomes can be appropriately and feasibly studied. Thought should be given in identifying the specific components of
the sedentary-behavior domain that should be measured. Using the definition of
Biddle et al12 that sedentary behavior is “a distinct class of behaviors characterized
by low energy expenditure,”(p 30) sedentary behaviors can include television watching, talking on the phone, reading, playing video or computer games, doing office
work, listening to music, driving or riding in a car, and using the computer. Each
of these activities can be performed in a variety of settings: at home, in offices,
at schools, while commuting, or in social settings. It is possible that regardless of
physical activity level, reducing sedentary behaviors in each of these settings could
provide health benefits, but the health outcomes of sedentary behavior need to be
documented more rigorously with improved measures. By specifically defining
the behaviors that might be included when assessing sedentary behaviors, more
targeted intervention programs and policy recommendations can be developed to
reduce sedentary behaviors.
Strengths and Limitations
Limitations of the study include the use of a convenience sample of participants
who were highly educated, resided in developed countries, and were not necessarily representative of the general population in that they might have provided more
reliable self-reports than other groups. The population examined contained some
students and individuals with university-based employment who might sit for longer
periods of time and might have different relationships between sedentary time
and physical activity than other populations. Future research will need to examine
the measurement properties of sedentary-behavior measures in more generalizable populations. The IPAQ was readministered a minimum of 3 days after initial
administration. There is a chance that some individuals remembered what they had
recorded in the initial visit, and this could have inflated test–retest reliability scores.
Partly because physical activity is overestimated by the IPAQ,47,48 few people were
classified as physically inactive, and the small cell sizes could have contributed to
the lack of association between sedentary behavior and inactivity. The validity of
using accelerometer counts <100 per minute as a measure of sedentary behavior is
not established, and further research, especially in adults, is warranted. However,
other objective measures of sedentary behavior are not available, and the cut point
of <100 counts per minute has some empirical support.39,43 The low sample sizes
could have affected validity correlations for the Netherlands, USA 1, and USA 2
samples. The brevity of the IPAQ sitting items might hinder accurate measurement of total sedentary behavior. It might be difficult for people to retrospectively
include all sedentary behavior without specific examples and prompting, resulting
in recall error. Spearman correlations were used to validate the IPAQ sitting items
IPAQ Sedentary Behavior S41
in the current study, and Spearman correlations do not take account of agreement
in the actual values; thus, it is possible to have high correlations and low agreement
between measures.
A main strength of the IPAQ sitting measure is that it is a brief measure of sedentary behavior that is suitable for use in large studies or for population monitoring
in which there is a paucity of accurate data on the prevalence of sedentary behavior.
An additional strength is that, to our knowledge, IPAQ sitting items are the only
sedentary-behavior measures with validity and reliability data for adults.
Conclusions
Valid and feasible measures that specifically assess sedentary behaviors rather than
a lack of activity are needed for monitoring sedentary behaviors in populations. One
population-monitoring survey that has begun to incorporate measures specific to
sedentary behavior is the US National Health and Nutrition Examination Survey,
which incorporates a measure of television and computer use and asks a general
question about whether activities during the day mostly involve sitting, standing
and walking, or more heavy work.49 It would be useful for surveillance questionnaires to also include sedentary measures that examine more than television and
computer use.
Being able to measure patterns of sedentary behavior can allow further research
into whether high volumes of sedentary behavior are harmful to health or whether
low amounts of sedentary behavior have protective effects on health. Developing
reliable and valid measures of sedentary behavior that cover the many domains of
sedentary behavior in 1 instrument would be useful for smaller-scale studies and
intervention research.
Rather than deducing sedentarism from instruments that measure physical activity, sedentary behaviors should be explicitly measured. The IPAQ was
developed as a surveillance measure of physical activity and included specific
items to assess sedentary behavior. The IPAQ sitting items have adequate reliability and validity for women and men from multiple countries and, thus, are
an advance on the current physical activity–surveillance methods for assessing
sedentary behavior. Using measures specific to sedentary behaviors will help
clarify the definitions of sedentary behavior and sedentary people, which have
been confused in previous research, and further elucidate the independent health
effects of sedentary behaviors. The term sedentary should be used to refer to
behaviors that are sedentary rather than people who are inactive. Defining people
as inactive based on a lack of leisure-time physical activity can be useful, but
care should be taken not to infer that inactive individuals engage in excessive
sedentary behavior.
References
1. World Health Organization. Diet, Nutrition, and the Prevention of Chronic Diseases:
Report of a Joint WHO/FAO Expert Consultation. Report No.: 916. Geneva, Switzerland:
World Health Organization; 2003.
2. Sturm R. The economics of physical activity: societal trends and rationales for interventions. Am J Prev Med. 2004;27(suppl 3):126-135.
S42 Rosenberg et al
3. Livingstone MBE, Robson PJ, Wallace JMW, McKinley MC. How active are we? levels
of routine physical activity in children and adults. Proc Nutr Soc. 2003;62:681-701.
4. Prentice AM, Jebb SA. Obesity in Britain: gluttony or sloth? Br Med J. 1995;311:437439.
5. Tudor-Locke C, Meyers AM. Challenges and opportunities for measuring physical
activity in sedentary adults. Sports Med. 2001;31:91-100.
6. US Department of Health and Human Services. Healthy People 2010: Understanding
and Improving Health. 2nd ed. Washington, DC: US Government Printing Office;
2000.
7. Bernstein MS, Morabia A, Sloutskis D. Definition and prevalence of sedentarism in an
urban population. Am J Public Health. 1999;89:862-867.
8. Varo JJ, Martinez-Gonzalez MA, Irala-Estevez J, Kearney J, Gibney MJ, Martinez
JA. Distribution and determinants of sedentary lifestyles in the European Union. Int J
Epidemiol. 2003;32:138-146.
9. Pratt M, Macera CA, Blanton C. Levels of physical activity and inactivity in children
and adults in the United States: current evidence and research issues. Med Sci Sports
Exerc. 1999;31(suppl 11):S526-S533.
10. Dietz W. The role of lifestyle in health: the epidemiology and consequences of inactivity. Proc Nutr Soc. 1996;55:829-840.
11. Robinson TN. Reducing children’s television viewing to prevent obesity: a randomized
controlled trial. J Am Med Assoc. 1999;282:1561-1567.
12. Biddle SJH, Gorely T, Marshall SJ, Murdey I, Cameron N. Physical activity and sedentary behaviors in youth: issues and controversies. J R Soc Health. 2004;124(1):2933.
13. Leslie E, Owen N, Salmon J, Bauman A, Sallis JF, Lo SK. Insufficiently active Australian college students: perceived personal, social, and environmental influences. Prev
Med. 1999;28(1):20-27.
14. Coon KA, Goldberg J, Rogers BL, Tucker KL. Relationships between use of television
during meals and children’s food consumption patterns. Pediatrics. 2001;107:e7.
15. Saelens BE, Sallis JF, Nader PR, Broyles SL, Berry CC, Taras HL. Home environmental
influences on children’s television watching from early to middle childhood. J Dev
Behav Pediatr. 2002;23(3):127-132.
16. Marshall SJ, Biddle SJH, Gorely T, Cameron N, Murdey I. Relationships between
media use, body fatness and physical activity in children and youth: meta-analysis. Int
J Obes. 2004;28:1238-1246.
17. Nelson MC, Gorden-Larsen P, Adair LS, Popkin BM. Adolescent physical ativity
and sedentary behavior: patterning and long-term maintenance. Am J Prev Med.
2005;28:259-266.
18. Marshall SJ, Biddle SJH, Sallis JF, McKenzie TL, Conway TL. Clustering of sedentary
behaviors and physical activity among youth: a cross-national study. Pediatr Exerc Sci.
2002;14:401-417.
19. Hu FB, Li TY, Colditz GA, Willett WC, Manson JE. Television watching and other
sedentary behaviors in relation to risk of obesity and type 2 diabetes mellitus in women.
J Am Med Assoc. 2003;289:1785-1791.
20. Brown WJ, Miller YD, Miller R. Sitting time and work patterns as indicators of overweight and obesity in Australian adults. Int J Obes. 2003;27:1340-1346.
21. Ching PLYH, Walter CW, Rimm EB, Colditz GA, Gortmaker SL, Stampfer MJ. Activity level and risk of overweight in male health professionals. Am J Public Health.
1996;86:25-30.
22. Vioque J, Torres A, Quiles J. Time spent watching television, sleep duration, and obesity
in adults living in Valencia, Spain. Int J Obes. 2000;24:1683-1688.
23. Tucker LA, Bagwell M. TV viewing and obesity in adult females. Am J Public Health.
1991;81:908-911.
IPAQ Sedentary Behavior S43
24. Tucker LA, Friedman GM. TV viewing and obesity in adult males. Am J Public Health.
1989;79:516-518.
25. Salmon J, Bauman AE, Crawford D, Timperio A, Owen N. The association between
television viewing and overweight among Australian adults participating in varying
levels of leisure time physical activity. Int J Obes. 2000;24:600-606.
26. Kronenberg F, Pereira MA, Schmitz KH, et al. Influence of leisure time PA and
television watching on atherosclerosis risk factors in the NHLBI family hear study.
Atherosclerosis. 2000;153:433-443.
27. Fitzgerald SJ, Kriska AM, Pereira MA, deCourten MP. Associations among physical
activity, television watching, and obesity in adult Pima Indians. Med Sci Sports Exerc.
1997;29:910-915.
28. Cameron AJ, Welborn TA, Zimmet PZ, et al. Overweight and obesity in Australia: the
1999–2000 Australian diabetes, obesity and lifestyle study. Med J Aust. 2003;9:427432.
29. Coakley EH, Rimm EB, Colditz G, Kawachi L, Willett W. Predictors of weight change
in men. Int J Obes. 1998;22:89-96.
30. Crawford D, Jeffery RW, French SA. Television viewing, physical inactivity, and obesity.
Int J Obes. 1999;23:437-440.
31. Sidney S, Sternfeld B, Haskell WL, Jacobs DR, Chesney MA, Hulley SB. Television
viewing and cardiovascular risk factors in young adults: the CARDIA study. Ann
Epidemiol. 1996;6:154-159.
32. Ball K, Brown WJ, Crawford D. Who does not gain weight? prevalence and predictors
of weight maintenance in young women. Int J Obes. 2002;26:1570-1578.
33. Martinez-Gonzalez MA, Martinez JA, Hu FB, Gibney MJ, Kearney J. Physical inactivity, sedentary lifestyle and obesity in the European Union. Int J Obes. 1999;23:11921201.
34. McCormack G, Giles-Corti B, Milligan R. The test–retest reliability of habitual incidental physical activity. Aust N Z J Public Health. 2003;27:428-433.
35. Macera CA, Jones DA, Yore MM, et al. Prevalence of physical activity, including
lifestyle activities among adults—United States, 2000–2001. Morb Mortal Wkly Rep.
2003;51:764-769.
36. Ham SA, Yore MM, Fulton JE, Kohl HW. Prevalence of no leisure-time PA—35 states
and the District of Columbia, 1988–2002. Morb Mortal Wkly Rep. 2004;53:82-86.
37. Kruger J, Ham SA, Kohl HW. Trends in leisure-time physical inactivity by age, sex,
and race/ethnicity—United States, 1994–2004. Morb Mortal Wkly Rep. 2005;54:991994.
38. Craig CL, Marshall AL, Sjostrom M, et al. International Physical Activity Questionnaire: 12 country reliability and validity. Med Sci Sports Exerc. 2003;35:1381-1395.
39. Ekelund U, Aman J, Yngve A, Renman C, Westerterp K, Sjostrom M. Physical activity
but not energy expenditure is reduced in obese adolescents: a case-control study. Am
J Clin Nutr. 2002;76:935-941.
40. Treuth MS, Schmitz KH, Catellier DJ, et al. Defining accelerometer thresholds for
activity intensities in adolescent girls. Med Sci Sports Exerc. 2004;36:1259-1266.
41. Puyau MR, Adolph AL, Vohra FA, Butte NF. Validation and calibration of physical
activity monitors in children. Obes Res. 2002;10:150-157.
42. Ekelund U, Yngve A, Sjostrom M, Westerterp K. Field evaluation of the Computer
Science and Application’s Inc. activity monitor during running and skating training in
adolescent athletes. Int J Sports Med. 2000;21:586-592.
43. Treuth MS, Hou N, Young DR, Maynard LM. Accelerometry-measured activity and sedentary time and overweight in rural boys and girls. Obes Res. 2005;13:1606-1614.
44. Swartz AM, Strath SJ, Bassett DR, O’Brien WL, King GA, Ainsworth BE. Estimation
of energy expenditure using CSA accelerometers at hip and wrist sites. Med Sci Sports
Exerc. 2000;32(suppl 9):S450-S456.
S44 Rosenberg et al
45. Matthews CE. Calibration of accelerometer output for adults. Med Sci Sports Exerc.
2005;37(suppl 11):S512-S522.
46. Jacobs DR, Ainsworth BE, Hartmann TJ, Leon AS. A simultaneous evaluation of 10
commonly used physical activity questionnaires. Med Sci Sports Exerc. 1993;25:8191.
47. Rzewnicki R, Auweele YV, De Bourdeaudhuij I. Addressing overreporting on the
International Physical Activity Questionnaire (IPAQ) telephone survey with a population sample. Public Health Nutr. 2003;6:299-305.
48. Johnson-Kozlow M, Sallis JF, Gilpin EA, Rock CL, Pierce JP. Comparative validation
of the IPAQ and the 7-day PAR among women diagnosed with breast cancer. Int J
Behav Nutr Phys Act. 2006;3:7.
49. US Department of Health and Human Services. National Health and Nutrition Examination Survey 2003–2004 Questionnaire. Hyattsville, MD: National Center for Health
Statistics; 2004.